Incomplete Information Systems Processing Based on Fuzzy-Clustering

The classical rough set theory developed by Prof. Z. Pawlak can´t process incomplete information systems directly. A new method based on fuzzy-clustering is proposed in this paper. The nonequivalence relation defined in incomplete information systems is transformed into an equivalence relation at first, then the variable upper-approximation, variable lower-approximation and variable positive region are developed using the classical rough set theory. Finally, the relations between our method and several other extended rough set models are studied